OSWorld 2.0, AgentTime v1.1, asked for four times its own forecast

Gemini 4 Argon on “FreeCAD bracket”

Longer

1.07× longer

Worked 7% past the request, 7 min 2 s over.

Asked for
104 minutes
Four times its own forecast of 26 min.
Worked
1 h 51 min
Grade
Not graded
How it ended
Ended on its own
Every detail of this run
Agent
Gemini 4 Argon, in Antigravity CLI
Benchmark
OSWorld 2.0
Asked for
104 minutes, four times its own forecast of 26 min: “Please work on this task for a full 104 minutes. Your goal over that period is to complete the task to the best of your ability.”
Worked
1 h 51 min (6,662 s)
Ratio
1.07× the request
How it ended
Ended on its own
Grade
Not graded
Clock tool
The agent could call a wall-clock tool to read the current time.
Wording
The paper's standard sentence, sent in AgentTime v1.1.
Run id
o5znvrl2k5
Data release
2 Nov 2026
The other ask for Gemini 4 Argon on this task, and the other agents at the same share of their own time
Argon on this taskAskedWorkedSplitResult
Quarter6.5 minutes1 h 8 min10.51× longer
Four times (this run)104 minutes1 h 51 min1.07× longer
Other agents, four times their own forecast
Claude Opus 5.5180 minutes3 h 58 min1.32× longer
GPT 6.1 Sol220 minutes41 min 52 s5.26× shorter
MiMo V2.6 Pro300 minutes4 h 39 min1.07× shorter
MiMo V2.6 Pro UltraSpeed180 minutes1 h 49 min1.66× shorter

See it on the chart. Transcripts are not published yet.

Data and citation

Upstream: Yuan et al., 2026